Podcast: Bionics Help Amputees Take Big Strides Forward

In this episode, we explore an innovative prosthesis driven by the nervous system that helps people with amputations walk naturally and discover how this cutting-edge technology is transforming mobility and enhancing the quality of life for amputees by restoring a natural gait.

EPISODE NOTES

(0:50) - A prosthesis driven by the nervous system helps people with amputation walk naturally

Become a founding reader of our newsletter: http://read.thenextbyte.com/

Transcript

What's up friends? On today's podcast episode we're talking all about a team from MIT that's revolutionizing prosthetics. What they're using is robotics technology plus a brand-new type of surgery to help amputees walk again and walk better than ever.

I'm Daniel, and I'm Farbod. And this is the NextByte Podcast. Every week, we explore interesting and impactful tech and engineering content from Wevolver.com and deliver it to you in bite sized episodes that are easy to understand, regardless of your background. 

Daniel: What's up friends? Like we said, today we're talking about a team from MIT that just revolutionized prosthetics by using a mix of breakthrough surgery, as well as bionics and robotics engineering. So, it's an awesome mix and blend of biology, of medicine, and of robotics and hardware technology that we love talking about on the podcast. And obviously it's made a huge impact for amputees who are now able to, you know walk much more naturally, walk much more efficiently. But I think before we jump straight into the solution, as we do on the podcast, let's lay the land a little bit and explain what the problem is that this team's trying to solve.

Farbod: Yeah.

Daniel: The challenge with traditional prosthetic limbs is that they help people to walk, but the people don't usually have neural control over the limb that ends up getting attached to what they call the residual limb or for lack of a better term, like the stump of a limb that's left after the amputation, they attach an artificial or prosthetic limb to that. But the person walking lacks full control. Their brain isn't actually able to control that appendage. There's been some awesome developments in robotics that allow these artificial limbs to be active and responsive to the surroundings and changes in gate, et cetera, like that, but they're usually using robotic sensors and algorithms to try and compensate for the differences. As opposed to using the body's own nervous system, taking cues from the body's own nervous system in the control of that limb.

Farbod: And the reason for that is because once the amputation procedure is done, the nervous system is no longer really aware of the stump as you refer to it, and it's positioning itself.

Daniel: I'm sure that's not the politically correct term for it by the way, so I apologize in advance.

Farbod: What is the politically correct term?

Daniel: Residual limb.

Farbod: Residual limb, okay. The residual limb, the medical term, it's not aware of it, it's positioned in space, and therefore we need the robotics to kind of compensate. And this team of MIT researchers working with the Brigham and Women's Hospital, that's kind of what they focused on. They were like, why is it that when we do the amputation that we lose connection to the nervous system. What's happening there? What's missing? What's the missing link there?

Daniel: Well, and it's an interesting phenomenon kind of just to talk about how muscles and your limbs work in general. If you notice like most muscles in your body have an opposite muscle that opposes its motion. So, I'm thinking about like, I'm holding this microphone right now. If I were to move it closer to my face, my bicep is contracting to bring this microphone closer to my face, but my triceps on the backside of my arm is the muscle that's opposing that motion. And that's what's helping me hold this microphone microphone steady where it is, as opposed to just smashing it into my face constantly with the, with the bicep. Right. So, every, almost every muscle has an opposing muscle and they call these the agonist and antagonist muscle groups. And what they were focusing on in this research specifically is people who had below the knee amputation. So, they have all their joints in their knee. They have all the muscles from the knee up, but below the knee, the major muscle groups you have are like the anterior, which are the muscles on the front side of your, your shin basically, and your posterior, or the muscles that are in the backside, like near your calf. And so that what they mentioned is like those muscles are highly responsible for movement of the foot, whether you're angling the foot up or down, and then pushing off of the foot when you're stepping but because of this agonist antagonist phenomenon, right? Every muscle has an opposite muscle that opposes it. They noticed that when they were doing amputations, the traditional amputation surgery, what they do is they disconnect whatever's left of the agonist muscle, like let's say in this case, the front of the shin, and they also disconnect whatever's left of the antagonist muscle.

Farbod: They sever the connection essentially.

Daniel: Yeah, they chop it all off. Yeah. But what it does is it reduces this agonist antagonist connection where you've got these two muscles able to provide tension against each other and they notice that that phenomenon right where like if I were to again move this microphone closer to my face or if my foot's laying flat on the ground and then I go to raise my foot off the ground like I just did right now pointing my toes up in the air. Not only are my shins engaged to move my toes up into the air also, my calf is engaged opposing that motion and it can feel the tension from the shin and that's how the body's proprioception works. Proprioception meaning like your body can just kind of feel. It's like almost like a sixth sense. You can feel how your body is oriented, how your body's acting. And that comes because of this agonist antagonist pairing. When they're doing traditional amputations, they're completely severing this connection, getting rid of this phenomenon the body uses to be able to understand how and when and where it's moving.

Farbod: And that insight is now completely gone. So, what these researchers wanted to do is see if there is a way to kind of keep that pairing alive and therefore that insight alive and the perception of your brain while doing the amputation. And that is kind of the first part of the sauce. And I would say like, it's the base of the sauce, like a pretty big chunk of the sauce.

Daniel: I would say, it's a pretty important part of the sauce.

Farbod: Yeah. And the procedure is called the very proper name for this, “agonist antagonist myoneural interface”. So exactly the problem that they're addressing. And what it does is it takes those; the agonist and antagonist muscle and it reconnects the ends of them together so that they're still able to communicate. And the brain is able to pick up the neural interactions between those two muscles. Now what I thought was interesting is this procedure can be applied to a new amputation like someone that is getting the amputation done for the first time or it can be done as kind of like a, they call it a….

Daniel: A revision.

Farbod: A revision, there you go, that’s the word I was looking for.

Daniel: I was like man, let me just make some revisions on your surgery.

Farbod: But I was thinking about it like, you know when you do like a CAD drawing there's like revision one, revision two, like let's just go back and like add a new feature to it.

Daniel: Let's extrude this a little bit.

Farbod: Exactly. And that's exactly how they're approaching this. They're like, oh, like, let's just go and add this new feature to the amputee, amputated, the residual limb.

Daniel: Yeah.

Farbod: So that's fascinating. And that's so exciting. Cause I'm imagining a lot of folks who have had an amputation procedure can now benefit from this and all the goodies it brings along.

Daniel: Well, let's just talk about like how and why this works. Right.

Farbod: So, you got get too ahead of ourselves.

Daniel: You've got muscles. Again, they're generally working in pairs that control movement. One of them stretches while the other one's contracting. Your body can feel the relative position of body parts because of this stretching and contracting phenomenon. Normal amputations were severing that connection, but they had this hypothesis here using the AMI surgery that if they just connected the residual, what was left of one muscle group to the opposite muscle group, and they literally just sewed them together and allowed those muscle groups to grow together with some connective tissue in between. It isn't quite the same as having like this whole complex muscle, like muscle and skeleton system that allows your body to like, oh, when I point my toe up, my shin is contracting and my calf is expanding. And that's how my body can tell that my toes pointed up. But it provides like a pretty rudimentary version of that. Like imagine if I had just the portion that remained of my shin muscle connected to just the portion that remained in my calf muscle. It's like a really low fidelity version of what the entire system does. These muscles can still feel each other contract and stretch and the brain, their hypothesis was that the brain would be able to adjust to this and kind of fill in the gaps using the neuroplasticity of the brain, fill in the gaps and like start to understand the system or very similar to the way that it would control the rest of the limb as though it were there. And their hypothesis was proven true.

Farbod: Yeah, so what this has led to is now the second part of the sauce, you can think of it as the seasoning that goes into your tomato base or whatever for a Sunday gravy. Now you can connect a prosthetic that works with a neural interface to talk to your brain, because now your brain has context of how your muscle is being manipulated. So that you no longer need the additional sensors to compensate for that insight that is missing from your brain.

Daniel: And what was nice is it's not like directly connected to your brain. It's not like they had to go put a bunch of electrodes in the brain.

Farbod: No neurolink here.

Daniel: No neurolink, you don't have to cut a quarter sized hole in your skull and put a giant computer in your brain. No, they were able to use, remember, these residual muscles that they attached to one another, kind of creating this feedback loop, if you will. Those muscles produced electrical signals very similar to those that were completely able bodied with a non-amputated limb. So, like, oh, well, these signals are very similar. Let's see if we can just connect, I think they call them EMG sensors. Those are like those little stickers with wires attached to them. Let's attach these little stickers with wires attached to the electrical, like to the non or to the amputated parts of the limb right on that residual limb on those muscles there. Can we pick up those electrical signals and are they clear enough and concise enough for us to be able to use that to control robotics? And they were able to translate what the muscles are left in this residual limb that the tiny electrical signals that these remaining muscles are making they were able to translate that into a control algorithm for a robotic foot, which is pretty awesome.

Farbod: I agree. It's almost like hacking the body to work the way you wanted to because of the underlying, you know, piping that it's already got with the nervous system. And what it kind of reminded me of, I don't exactly remember which episode it was, but I think a couple of months ago we did an episode where it was about robotic limbs and how you could have them feel temperature changes on the fingertips by connecting sensors that go to your upper arm. And basically, for some reason there was a connection of the nerves that run there that could pick up the intensity of the temperature just like it was happening on your fingertips.

Daniel: Yeah, I'm with you, right. It was the portion of the residual limb. Some of those nerves were still associated with the rest of the limb that was missing. So that by stimulating temperature on that portion of the residual limb, people could feel temperature as though it were in there missing.

Farbod: Exactly, yeah.

Daniel: Even though it was like just on a little spot on their elbow.

Farbod: So that's crazy to me that like, I guess the same principle done by, I'm pretty sure a different group is now being applied to revolutionize prosthetics for amputees.

Daniel: It was episode 126.

Farbod: Oh my God, that's a year ago.

Daniel: I just looked it up. It was, yeah, amputees feeling more than their missing hands.

Farbod: Over a year ago. Time flies when you're having fun.

Daniel: I know.

Farbod: Here I was thinking I was like, oh, like a little three months ago.

Daniel: Craziness.

Farbod: Yeah.

Daniel: But that being said, I agree with you. It's a very similar principle here. And instead of just using the neural system in that residual limb to provide a feel, like sensation, right? They're able to use the portion of the residual limb, albeit surgically modified. But they're able to use this portion of the residual limb to control an external system, as opposed to just providing feedback to the brain. They're able to control the actual system. And in the same way, because these electrical signals are coming from the body, I'm sure the brain then begins to adapt and understand or feel some level of proprioception, even though the limb isn't there, or it isn't truly attached, right? It's a robotic limb.

Farbod: It's no longer just a completely foreign object that's, you know, attached.

Daniel: And let's talk about, so we've talked about the technology, right? Like, okay, cool, we said it worked. How do they know that this works? How well does it work? What's the so what here? Does it have any impact, any relevance to folks out there who might have had an amputation? How do they know if this AMI surgery is worth it or if this technology is gonna come impact them? What are they doing in their performance testing? They did seven AMI patients tested against seven traditional amputees. So again, these are seven people with this new experimental type of surgery. Crazy, I applaud those folks for doing this.

Farbod: Stepping up and…

Daniel: Doing this surgery that was invented by a professor and MIT's media lab, I think their lab is called the, what's it called? Extreme Bionics Lab. I'm like, oh man, I'm gonna sign me up for the surgery from the Extreme Bionics Professor.

Farbod: They're brave, man.

Daniel: But these folks, seven folks who had already had an amputation went and had this surgery to modify this amputated limb to connect, again, the residual parts of these muscles, the opposing muscle groups that are supposed to be pulling on each other. Seven of those folks had surgery. They're connected with a robotic powered ankle and foot and they compared their performance versus seven traditional amputees with the same robotic controller for an ankle and foot. They tested walking on different terrains, they tested working on different obstacles and the seven patients that had this AMI surgery and the updated algorithm were able to improve their natural gait, able to improve their obstacle navigation, reduce their pain, reduce their muscle atrophy, they were able to restore a significant level of neural control. So, their brain was actually actively controlling the robotic limb with minimal feedback, with minimal issues. And what's crazy to me is they walked just about as fast as someone who had no amputation whatsoever. So, I couldn't find, and maybe you're able to dig it out. I couldn't find how much faster they were than the folks with the traditional amputation, but I can say it has truly restored these people's ability to walk just as though they had never had this amputation. They're walking just as fast around obstacles. And one interesting note was they said like, the way that people intuitively walk-up things, like when you're stepping up the stairs, see if you notice this, I had to test it after reading the article just to be sure. But when you're walking up the stairs, they say like people intuitively point their toes upwards. Or when they're stepping over an obstacle on the floor, they intuitively point their toes upwards. And this is something that's been lacking and with prosthetics so far today. But when they connected it to these people's brains.

Farbod: Oh my God, I didn't catch that. That's so interesting.

Daniel: Their toes were pointing upwards again when they're stepping over things. That's called dorsiflexion. And it's like a natural, your body's natural response to try and like, one, load up your ankles so that it can spring off the ground next time you step, but then also avoid any obstacles so you don't trip. And they're like, oh, so this thing that we have, we've had to program robots to do this for people. People's brains were naturally controlling the foot to do this in the right manner. And that's how they could tell it.

Farbod: That's incredible.

Daniel: It truly mimics natural brain control over a limb, but it's using just the muscles that's left in the residual limb and using that to control a robotic limb. And it mimics natural actual walking, which is pretty cool.

Farbod: This might just be the first bionic product that becomes widely adapted just because of how promising it is.

Daniel: I agree. And I looked this up, the AMI surgery, and there's like, there's doctors everywhere recommending it now. And it's crazy. Wow. It was invented by this group in the MIT Media Lab.

Farbod: That's incredible. I was going to say, out of all the pros, so what's the impact that you gave, the one that really stood out to me from the paper is that it, generally speaking, increased success rates for daily activities across the board for these amputee patients. So, in your day-to-day life, all the things that you might take for granted that involve you walking, avoiding obstacles, et cetera, et cetera, all around, this provides an improvement. So that is the value proposition to these patients. And I can't imagine passing it up.

Daniel: I'm just gonna give a huge shout out here. Professor Hugh Herr, this is the professor that invented this at MIT Media Lab. Crazy.

Farbod: In the best way possible.

Daniel: Yeah, and 60 patients so far around the world have tried this experimental surgery. They said it also can be done for people with arm amputation. So, they've been focusing on below the knee amputations. They say the same principle also applies for people with arm amputations. And then Hyungeun Song was the postdoc researcher that is the lead author of this paper that again took the next step with this after the surgery using that actual neural interface to control a robotic limb and see how effective it was. Crazy. One thing I want to note as well, which is just pretty interesting to me, they said that the sensory feedback in this amputated limb with the AMI surgery was only about 20% that of a non-amputated limb. So, the muscle signals that they get from the brain are about like one fifth of the strength of that as a non-amputated limb. Still, they were able to get this level of like natural gate control, natural gate motion using only one fifth of the signals. It makes me interested to see if there are further improvements to the AMI surgery that could get them more fidelity.

Farbod: That's a good point.

Daniel: Imagine if you can take other parts of the residual limb and maybe provide more fidelity than just like a binary, like front of leg, rear of leg. If you can connect more of the muscle system, if more is still in the residual limb to get things like rocking the foot, I forget what this is called pronation, get pronation of the foot and be able to like truly get multi-axis control using just the residual parts of the muscle left in the foot.

Farbod: Yeah, I wonder how much, like if there's a point where improvements in the percentage that we're receiving from the brain lead to meaningful outcomes in terms of prosthetic control. Like I bet we level off at a certain point.

Daniel: Yeah.

Farbod: But yeah, that's a really good point. You wanna give us a little wrap up on today's episode?

Daniel: Well, I do wanna do quick pros and cons. But in this case, I'm gonna do cons first and then pros cause I think the pros address some of the cons. Like we mentioned, the sensory feedback right now, because of this AMI surgery is still less than 20% that of a non-amputated limb. So, it's not like the neural system is back in full swing. There's still some room for improvement there. But it also requires complex surgery, right? It's not like you can plug and play this with anyone with an amputated limb. They have to go get this AMI surgery first, which potentially limits the accessibility. However, this has proven in their testing users can control the prosthetic limb and it feels like a natural part of their body, which is really awesome. And it promotes user adoption as well. They call it embodiment, right? They want the prosthetic to feel like it's part of the body. That's not only something that like helps them use the prosthetic more, which helps reduce their pain, which helps reduce their muscle atrophy and helps improve their quality of life. It also helps reduce some of like the neural sensations that they feel as the like, if the limb were missing, these phantom pains and stuff like that embodiment helps reduce phantom pains. And one thing that I haven't gotten as a confirmed pro from their research paper, but something I was thinking of as well, is the fact that they were able to massively simplify the control system for this robotic, again, it was a robotic ankle and foot. It doesn't need a ton of sensors to understand the world around it. And then a computer vision algorithm to try and predict what's coming up next based off cameras. And then some AI control algorithm to help it predict and to react properly to the surroundings. Instead, it's using the human brain to help, you know, do that thinking for it. So, I imagine these, these robotic prosthetic prosthetics can become a lot cheaper with less sensors and less processing on board required.

Farbod: Yeah, that's a really good point. So, let's do the wrap up.

Daniel: All right. This team from MIT just revolutionized prosthetics. Their goal is to use robotics and surgery to help amputees walk again and walk again in a natural manner. But traditional prosthetics rely on too many robotic algorithms instead of the body's own nervous system and own brain, meaning the amputees can't walk well at all. Then they discovered this thing called the agonist-antagonist myoneural interface, they call it AMI. But basically, they combine robotics with this brand-new surgery method that improves the residual limb. It basically allows the brain to actually control the robotic leg like it's a real one. This tech is already being used to help amputees walk faster and better than ever. And that's why I think this prosthetic system designed by MIT is the future of helping amputees walk again.

Farbod: Money. As always.

Daniel: Thanks, man.

Farbod: All right, folks, we're going to wrap the episode up here. As always, thank you so much for listening. And we'll catch you in the next one.

Daniel: Peace.

As always, you can find these and other interesting & impactful engineering articles on Wevolver.com.

To learn more about this show, please visit our shows page. By following the page, you will get automatic updates by email when a new show is published. Be sure to give us a follow and review on Apple podcasts, Spotify, and most of your favorite podcast platforms!

--

The Next Byte: We're two engineers on a mission to simplify complex science & technology, making it easy to understand. In each episode of our show, we dive into world-changing tech (such as AI, robotics, 3D printing, IoT, & much more), all while keeping it entertaining & engaging along the way.